# General imports 
import os 
import re
import numpy as np
import pandas as pd
from pathlib import Path
import scipy.stats as stats 
from skimage import measure
import matplotlib.pyplot as plt

# Pynwb imports
from hdmf_zarr import NWBZarrIO
from nwbwidgets import nwb2widget

data directory set to /data/

Index(['project_name', 'session_type', '_id', 'name', 'subject_id', 'genotype',
       'virus', 'date_of_birth', 'sex', 'modality', 'session_date', 'age',
       'session_time', 'targeted_structure', 'ophys_fov', 'session_number'],
      dtype='object')

array(['Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-ICL-ChRmine-oScarlet-IRES2-tTA2-WPRE)-hyg/wt',
       'Camk2a-tTA/wt;tetO-GCaMP6s/wt',
       'Camk2a-tTA/wt;TetO-jGCaMP8s-01/wt'], dtype=object)

subject_id
731015        7
740369        7
754303        6
767715        6
772414        4
766719        3
Name: count, dtype: int64

Selected subject_id is 731015

Selected session is single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

/opt/conda/lib/python3.10/site-packages/hdmf/spec/namespace.py:535: UserWarning: Ignoring cached namespace 'core' version 2.7.0 because version 2.8.0 is already loaded.
  warn("Ignoring cached namespace '%s' version %s because version %s is already loaded."

# General imports 
import os 
import re
import numpy as np
import pandas as pd
from pathlib import Path
import scipy.stats as stats 
from skimage import measure
import matplotlib.pyplot as plt

# Pynwb imports
from hdmf_zarr import NWBZarrIO
from nwbwidgets import nwb2widget

data directory set to /data/

Index(['project_name', 'session_type', '_id', 'name', 'subject_id', 'genotype',
       'virus', 'date_of_birth', 'sex', 'modality', 'session_date', 'age',
       'session_time', 'targeted_structure', 'ophys_fov', 'session_number'],
      dtype='object')

array(['Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-ICL-ChRmine-oScarlet-IRES2-tTA2-WPRE)-hyg/wt',
       'Camk2a-tTA/wt;tetO-GCaMP6s/wt',
       'Camk2a-tTA/wt;TetO-jGCaMP8s-01/wt'], dtype=object)

subject_id
731015        7
740369        7
754303        6
767715        6
772414        4
766719        3
Name: count, dtype: int64

Selected subject_id is 731015

Selected session is single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

/opt/conda/lib/python3.10/site-packages/hdmf/spec/namespace.py:535: UserWarning: Ignoring cached namespace 'core' version 2.7.0 because version 2.8.0 is already loaded.
  warn("Ignoring cached namespace '%s' version %s because version %s is already loaded."

# General imports 
import os 
import re
import numpy as np
import pandas as pd
from pathlib import Path
import scipy.stats as stats 
from skimage import measure
import matplotlib.pyplot as plt

# Pynwb imports
from hdmf_zarr import NWBZarrIO
from nwbwidgets import nwb2widget

# set data path
import sys
import platform
from os.path import join as pjoin

platstring = platform.platform()
system = platform.system()
if system == "Darwin":
    # macOS
    data_dir = "/Volumes/Brain2025/"
elif system == "Windows":
    # Windows (replace with the drive letter of USB drive)
    data_dir = "E:/"
elif "amzn" in platstring:
    # then on CodeOcean
    data_dir = "/data/"
else:
    # then your own linux platform
    # EDIT location where you mounted hard drive
    data_dir = "/media/$USERNAME/Brain2025/"
    
print('data directory set to', data_dir)

data directory set to /data/

metadata = pd.read_csv(os.path.join(data_dir, 'metadata', 'bci_metadata.csv'))
metadata.head()

Index(['project_name', 'session_type', '_id', 'name', 'subject_id', 'genotype',
       'virus', 'date_of_birth', 'sex', 'modality', 'session_date', 'age',
       'session_time', 'targeted_structure', 'ophys_fov', 'session_number'],
      dtype='object')

array(['Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-ICL-ChRmine-oScarlet-IRES2-tTA2-WPRE)-hyg/wt',
       'Camk2a-tTA/wt;tetO-GCaMP6s/wt',
       'Camk2a-tTA/wt;TetO-jGCaMP8s-01/wt'], dtype=object)

subject_id
731015        7
740369        7
754303        6
767715        6
772414        4
766719        3
Name: count, dtype: int64

Selected subject_id is 731015

Selected session is single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

/opt/conda/lib/python3.10/site-packages/hdmf/spec/namespace.py:535: UserWarning: Ignoring cached namespace 'core' version 2.7.0 because version 2.8.0 is already loaded.
  warn("Ignoring cached namespace '%s' version %s because version %s is already loaded."

Loaded NWB file from: /data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

metadata = pd.read_csv(os.path.join(data_dir, 'metadata', 'bci_metadata.csv'))
metadata.head()

Index(['project_name', 'session_type', '_id', 'name', 'subject_id', 'genotype',
       'virus', 'date_of_birth', 'sex', 'modality', 'session_date', 'age',
       'session_time', 'targeted_structure', 'ophys_fov', 'session_number'],
      dtype='object')

array(['Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-ICL-ChRmine-oScarlet-IRES2-tTA2-WPRE)-hyg/wt',
       'Camk2a-tTA/wt;tetO-GCaMP6s/wt',
       'Camk2a-tTA/wt;TetO-jGCaMP8s-01/wt'], dtype=object)

subject_id
731015        7
740369        7
754303        6
767715        6
772414        4
766719        3
Name: count, dtype: int64

Selected subject_id is 731015

Selected session is single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

/opt/conda/lib/python3.10/site-packages/hdmf/spec/namespace.py:535: UserWarning: Ignoring cached namespace 'core' version 2.7.0 because version 2.8.0 is already loaded.
  warn("Ignoring cached namespace '%s' version %s because version %s is already loaded."

Loaded NWB file from: /data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

metadata = pd.read_csv(os.path.join(data_dir, 'metadata', 'bci_metadata.csv'))
metadata.head()

metadata.columns

Index(['project_name', 'session_type', '_id', 'name', 'subject_id', 'genotype',
       'virus', 'date_of_birth', 'sex', 'modality', 'session_date', 'age',
       'session_time', 'targeted_structure', 'ophys_fov', 'session_number'],
      dtype='object')

metadata.genotype.unique()

array(['Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-ICL-ChRmine-oScarlet-IRES2-tTA2-WPRE)-hyg/wt',
       'Camk2a-tTA/wt;tetO-GCaMP6s/wt',
       'Camk2a-tTA/wt;TetO-jGCaMP8s-01/wt'], dtype=object)

metadata.value_counts(['subject_id'])

subject_id
731015        7
740369        7
754303        6
767715        6
772414        4
766719        3
Name: count, dtype: int64

# Pick the first mouse
subject_ids = np.sort(metadata['subject_id'].unique())
subject_id = subject_ids[0]
print('Selected subject_id is', subject_id)

# Look at its metadata, sorted by 'session_number'
this_mouse_metadata = metadata[metadata['subject_id']==subject_id].sort_values(by='session_number')
this_mouse_metadata

Selected subject_id is 731015

Selected session is single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

/opt/conda/lib/python3.10/site-packages/hdmf/spec/namespace.py:535: UserWarning: Ignoring cached namespace 'core' version 2.7.0 because version 2.8.0 is already loaded.
  warn("Ignoring cached namespace '%s' version %s because version %s is already loaded."

Loaded NWB file from: /data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

VBox(children=(HBox(children=(Label(value='session_description:', layout=Layout(max_height='40px', max_width='…

dff shape (nframes, nrois): (249370, 1736)

Frame Rate: 58.2634

# Pick the first mouse
subject_ids = np.sort(metadata['subject_id'].unique())
subject_id = subject_ids[0]
print('Selected subject_id is', subject_id)

# Look at its metadata, sorted by 'session_number'
this_mouse_metadata = metadata[metadata['subject_id']==subject_id].sort_values(by='session_number')
this_mouse_metadata

Selected subject_id is 731015

Selected session is single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

/opt/conda/lib/python3.10/site-packages/hdmf/spec/namespace.py:535: UserWarning: Ignoring cached namespace 'core' version 2.7.0 because version 2.8.0 is already loaded.
  warn("Ignoring cached namespace '%s' version %s because version %s is already loaded."

Loaded NWB file from: /data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

VBox(children=(HBox(children=(Label(value='session_description:', layout=Layout(max_height='40px', max_width='…

dff shape (nframes, nrois): (249370, 1736)

Frame Rate: 58.2634

# Pick the first mouse
subject_ids = np.sort(metadata['subject_id'].unique())
subject_id = subject_ids[0]
print('Selected subject_id is', subject_id)

# Look at its metadata, sorted by 'session_number'
this_mouse_metadata = metadata[metadata['subject_id']==subject_id].sort_values(by='session_number')
this_mouse_metadata

Selected subject_id is 731015

# Let's pick the last session for this mouse
session_name = this_mouse_metadata.name.values[-1]
print('Selected session is', session_name)

Selected session is single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

# Get the directory for this dataset
bci_data_dir = os.path.join(data_dir, 'brain-computer-interface')
print(bci_data_dir)

/data/brain-computer-interface

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

/opt/conda/lib/python3.10/site-packages/hdmf/spec/namespace.py:535: UserWarning: Ignoring cached namespace 'core' version 2.7.0 because version 2.8.0 is already loaded.
  warn("Ignoring cached namespace '%s' version %s because version %s is already loaded."

Loaded NWB file from: /data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

VBox(children=(HBox(children=(Label(value='session_description:', layout=Layout(max_height='40px', max_width='…

dff shape (nframes, nrois): (249370, 1736)

Frame Rate: 58.2634

/tmp/ipykernel_1392/2577709309.py:13: MatplotlibDeprecationWarning: The get_cmap function was deprecated in Matplotlib 3.7 and will be removed in 3.11. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap()`` or ``pyplot.get_cmap()`` instead.
  colors = cm.get_cmap('Paired')

dict_keys(['PhotostimTrials', 'Trials'])

# Get the directory for this dataset
bci_data_dir = os.path.join(data_dir, 'brain-computer-interface')
print(bci_data_dir)

/data/brain-computer-interface

# Get the data folder for this session
session_dir = os.path.join(bci_data_dir, session_name)
print(session_dir)

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58

# Now find the NWB file and set the path to load it
nwb_file = [file for file in os.listdir(session_dir) if 'nwb' in file][0]
nwb_path = os.path.join(session_dir, nwb_file)
print(nwb_path)

/data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

with NWBZarrIO(str(nwb_path), 'r') as io:
    nwbfile = io.read()
    print('Loaded NWB file from:', nwb_path)

/opt/conda/lib/python3.10/site-packages/hdmf/spec/namespace.py:535: UserWarning: Ignoring cached namespace 'core' version 2.7.0 because version 2.8.0 is already loaded.
  warn("Ignoring cached namespace '%s' version %s because version %s is already loaded."

Loaded NWB file from: /data/brain-computer-interface/single-plane-ophys_731015_2025-01-31_20-37-19_processed_2025-08-06_05-37-58/single-plane-ophys_731015_2025-01-31_20-37-19_behavior_nwb

nwbfile

nwb2widget(nwbfile)

VBox(children=(HBox(children=(Label(value='session_description:', layout=Layout(max_height='40px', max_width='…

dff shape (nframes, nrois): (249370, 1736)

Frame Rate: 58.2634

/tmp/ipykernel_1392/2577709309.py:13: MatplotlibDeprecationWarning: The get_cmap function was deprecated in Matplotlib 3.7 and will be removed in 3.11. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap()`` or ``pyplot.get_cmap()`` instead.
  colors = cm.get_cmap('Paired')

dict_keys(['PhotostimTrials', 'Trials'])

array(['photostim', 'photostim_post'], dtype=object)

Index(['start_time', 'stop_time', 'go_cue', 'hit', 'lick_L', 'reward_time',
       'threshold_crossing_times', 'zaber_step_times', 'tiff_file',
       'start_frame', 'stop_frame', 'conditioned_neuron_x',
       'conditioned_neuron_y', 'closest_roi'],
      dtype='object')

array([124])

spontaneous pre 2P frames 13799
spontaneous post 2P frames 15528

(249370, 1736)

nwb2widget(nwbfile)

VBox(children=(HBox(children=(Label(value='session_description:', layout=Layout(max_height='40px', max_width='…

dff shape (nframes, nrois): (249370, 1736)

Frame Rate: 58.2634

/tmp/ipykernel_1392/2577709309.py:13: MatplotlibDeprecationWarning: The get_cmap function was deprecated in Matplotlib 3.7 and will be removed in 3.11. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap()`` or ``pyplot.get_cmap()`` instead.
  colors = cm.get_cmap('Paired')

dict_keys(['PhotostimTrials', 'Trials'])

array(['photostim', 'photostim_post'], dtype=object)

Index(['start_time', 'stop_time', 'go_cue', 'hit', 'lick_L', 'reward_time',
       'threshold_crossing_times', 'zaber_step_times', 'tiff_file',
       'start_frame', 'stop_frame', 'conditioned_neuron_x',
       'conditioned_neuron_y', 'closest_roi'],
      dtype='object')

array([124])

spontaneous pre 2P frames 13799
spontaneous post 2P frames 15528

(249370, 1736)

(1736, 13799)

dff shape (nframes, nrois): (249370, 1736)

Frame Rate: 58.2634

/tmp/ipykernel_1392/2577709309.py:13: MatplotlibDeprecationWarning: The get_cmap function was deprecated in Matplotlib 3.7 and will be removed in 3.11. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap()`` or ``pyplot.get_cmap()`` instead.
  colors = cm.get_cmap('Paired')

dict_keys(['PhotostimTrials', 'Trials'])

array(['photostim', 'photostim_post'], dtype=object)

Index(['start_time', 'stop_time', 'go_cue', 'hit', 'lick_L', 'reward_time',
       'threshold_crossing_times', 'zaber_step_times', 'tiff_file',
       'start_frame', 'stop_frame', 'conditioned_neuron_x',
       'conditioned_neuron_y', 'closest_roi'],
      dtype='object')

array([124])

spontaneous pre 2P frames 13799
spontaneous post 2P frames 15528

(249370, 1736)

(1736, 13799)

dff_traces = nwbfile.processing["processed"].data_interfaces["dff"].roi_response_series["dff"].data
print('dff shape (nframes, nrois):',np.shape(dff_traces))

dff shape (nframes, nrois): (249370, 1736)

frame_rate = nwbfile.imaging_planes["processed"].imaging_rate
print('Frame Rate:', frame_rate)

Frame Rate: 58.2634

# Pick an ROI
ROI = 100

# Plot dff trace for selected ROI 
plt.rcParams["figure.figsize"] = (10, 4)
plt.plot(dff_traces[:, ROI], label=f'ROI {ROI}', color='black')

# Add shaded regions for stimulus epochs 
epoch_table = nwbfile.intervals["epochs"].to_dataframe()
stimulus_names = epoch_table.stimulus_name.unique()

import matplotlib.cm as cm
colors = cm.get_cmap('Paired')
colors = colors(np.linspace(0, 1, len(stimulus_names)))

                
for c, stimulus_name in enumerate(stimulus_names):
    stim_epoch = epoch_table[epoch_table.stimulus_name==stimulus_name]
    for j in range(len(stim_epoch)):
        plt.axvspan(xmin=stim_epoch.start_frame.iloc[j], xmax=stim_epoch.stop_frame.iloc[j], color=colors[c], alpha=0.3, label=stimulus_name)

plt.ylabel('dF/F')
plt.xlabel('2P Frame')
plt.title('Stimulus epochs and dF/F for ROI '+str(ROI))
plt.legend(bbox_to_anchor=(1.0, 1.0))
plt.show()

/tmp/ipykernel_1392/2577709309.py:13: MatplotlibDeprecationWarning: The get_cmap function was deprecated in Matplotlib 3.7 and will be removed in 3.11. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap()`` or ``pyplot.get_cmap()`` instead.
  colors = cm.get_cmap('Paired')

# These are the available trials tables
nwbfile.stimulus.keys()

dict_keys(['PhotostimTrials', 'Trials'])

array(['photostim', 'photostim_post'], dtype=object)

Index(['start_time', 'stop_time', 'go_cue', 'hit', 'lick_L', 'reward_time',
       'threshold_crossing_times', 'zaber_step_times', 'tiff_file',
       'start_frame', 'stop_frame', 'conditioned_neuron_x',
       'conditioned_neuron_y', 'closest_roi'],
      dtype='object')

array([124])

spontaneous pre 2P frames 13799
spontaneous post 2P frames 15528

(249370, 1736)

(1736, 13799)

(1736, 15528)

Text(0.5, 1.0, 'Spontaneous pre')

Text(0.5, 1.0, 'Spontaneous post')

# These are the available trials tables
nwbfile.stimulus.keys()

dict_keys(['PhotostimTrials', 'Trials'])

# Check the epoch table - which epochs have trials? 
epoch_table

photostim = nwbfile.stimulus["PhotostimTrials"].to_dataframe()
photostim.head()

array(['photostim', 'photostim_post'], dtype=object)

Index(['start_time', 'stop_time', 'go_cue', 'hit', 'lick_L', 'reward_time',
       'threshold_crossing_times', 'zaber_step_times', 'tiff_file',
       'start_frame', 'stop_frame', 'conditioned_neuron_x',
       'conditioned_neuron_y', 'closest_roi'],
      dtype='object')

array([124])

spontaneous pre 2P frames 13799
spontaneous post 2P frames 15528

(249370, 1736)

(1736, 13799)

(1736, 15528)

Text(0.5, 1.0, 'Spontaneous pre')

Text(0.5, 1.0, 'Spontaneous post')

there are 226 ROIs with soma_probability > 0.005

photostim = nwbfile.stimulus["PhotostimTrials"].to_dataframe()
photostim.head()

# Stimuli in photostim trials table
photostim['stimulus_name'].unique()

array(['photostim', 'photostim_post'], dtype=object)

# Compare to epochs
epoch_table

# Load the BCI trials table
trials = nwbfile.stimulus["Trials"].to_dataframe()
trials.head()

Index(['start_time', 'stop_time', 'go_cue', 'hit', 'lick_L', 'reward_time',
       'threshold_crossing_times', 'zaber_step_times', 'tiff_file',
       'start_frame', 'stop_frame', 'conditioned_neuron_x',
       'conditioned_neuron_y', 'closest_roi'],
      dtype='object')

array([124])

spontaneous pre 2P frames 13799
spontaneous post 2P frames 15528

(249370, 1736)

(1736, 13799)

(1736, 15528)

Text(0.5, 1.0, 'Spontaneous pre')

Text(0.5, 1.0, 'Spontaneous post')

there are 226 ROIs with soma_probability > 0.005

(256, 512)

# Load the BCI trials table
trials = nwbfile.stimulus["Trials"].to_dataframe()
trials.head()

# What info do we have for each trial?
trials.columns

Index(['start_time', 'stop_time', 'go_cue', 'hit', 'lick_L', 'reward_time',
       'threshold_crossing_times', 'zaber_step_times', 'tiff_file',
       'start_frame', 'stop_frame', 'conditioned_neuron_x',
       'conditioned_neuron_y', 'closest_roi'],
      dtype='object')

# The conditioned neuron ID is indicated by the 'closest_roi' column
trials.closest_roi.unique()

array([124])

# Load the epochs table from the NWB file
epoch_table = nwbfile.intervals["epochs"].to_dataframe()
epoch_table

spontaneous pre 2P frames 13799
spontaneous post 2P frames 15528

(249370, 1736)

(1736, 13799)

(1736, 15528)

Text(0.5, 1.0, 'Spontaneous pre')

Text(0.5, 1.0, 'Spontaneous post')

there are 226 ROIs with soma_probability > 0.005

(256, 512)

(1736, 13799)

(226, 13799)

# Load the epochs table from the NWB file
epoch_table = nwbfile.intervals["epochs"].to_dataframe()
epoch_table

# Compute the duration as difference between stop and start time
epoch_table['duration'] = epoch_table.stop_time-epoch_table.start_time
epoch_table

# Get just the spontaneous epochs
spontaneous_epochs = epoch_table[epoch_table.stimulus_name.str.contains('spont')]
spontaneous_epochs

spontaneous pre 2P frames 13799
spontaneous post 2P frames 15528

(249370, 1736)

(1736, 13799)

(1736, 15528)

Text(0.5, 1.0, 'Spontaneous pre')

Text(0.5, 1.0, 'Spontaneous post')

there are 226 ROIs with soma_probability > 0.005

(256, 512)

(1736, 13799)

(226, 13799)

# Get just the spontaneous epochs
spontaneous_epochs = epoch_table[epoch_table.stimulus_name.str.contains('spont')]
spontaneous_epochs

# Now get the start and stop times for the spontaneous epochs before and after the BCI task

# The spontaneous period before the task consists of the first two spontaneous blocks
spontaneous_pre_start = spontaneous_epochs.loc[1].start_frame
spontaneous_pre_stop = spontaneous_epochs.loc[2].stop_frame

# The last spontaneous period is the last spontaneous epoch
spontaneous_post_start = spontaneous_epochs.loc[4].start_frame
spontaneous_post_stop = spontaneous_epochs.loc[4].stop_frame

# Now you have the start and end frames for each epoch and can compute the duration of each
print('spontaneous pre 2P frames', spontaneous_pre_stop-spontaneous_pre_start)
print('spontaneous post 2P frames', spontaneous_post_stop-spontaneous_post_start)

spontaneous pre 2P frames 13799
spontaneous post 2P frames 15528

# Get the dFF traces from the NWB file
dff_traces = nwbfile.processing["processed"].data_interfaces["dff"].roi_response_series["dff"].data
# Look at the shape - the larger dimension are the 2P frames
print(dff_traces.shape)

(249370, 1736)

(1736, 13799)

(1736, 15528)

Text(0.5, 1.0, 'Spontaneous pre')

Text(0.5, 1.0, 'Spontaneous post')

there are 226 ROIs with soma_probability > 0.005

(256, 512)

(1736, 13799)

(226, 13799)

(226, 13799)

# Get the dFF traces from the NWB file
dff_traces = nwbfile.processing["processed"].data_interfaces["dff"].roi_response_series["dff"].data
# Look at the shape - the larger dimension are the 2P frames
print(dff_traces.shape)

(249370, 1736)

# Index into the dff traces array to limit to the relevant time points
dff_pre = dff_traces[spontaneous_pre_start:spontaneous_pre_stop, :]
# Now transpose so rows are ROI IDs
dff_pre = dff_pre.T
print(dff_pre.shape)

(1736, 13799)

(1736, 15528)

Text(0.5, 1.0, 'Spontaneous pre')

Text(0.5, 1.0, 'Spontaneous post')

there are 226 ROIs with soma_probability > 0.005

(256, 512)

(1736, 13799)

(226, 13799)

(226, 13799)

Text(0.5, 1.0, 'Spontaneous pre')

# Index into the dff traces array to limit to the relevant time points
dff_pre = dff_traces[spontaneous_pre_start:spontaneous_pre_stop, :]
# Now transpose so rows are ROI IDs
dff_pre = dff_pre.T
print(dff_pre.shape)

(1736, 13799)

# Index into the dff traces array to limit to the relevant time points
dff_post = dff_traces[spontaneous_post_start:spontaneous_post_stop, :]
# Now transpose so rows are ROI IDs
dff_post = dff_post.T
print(dff_post.shape)

(1736, 15528)

plt.pcolormesh(dff_pre, vmin=0, vmax=5, cmap='magma')
plt.colorbar(label='dFF')
plt.xlabel('2P frames')
plt.ylabel('ROI ID')
plt.title('Spontaneous pre')

Text(0.5, 1.0, 'Spontaneous pre')

plt.pcolormesh(dff_post, vmin=0, vmax=5, cmap='magma')
plt.colorbar(label='dFF')
plt.xlabel('2P frames')
plt.ylabel('ROI ID')
plt.title('Spontaneous post')

Text(0.5, 1.0, 'Spontaneous post')

# Get the roi_table from the NWB file
roi_table = nwbfile.processing["processed"].data_interfaces["image_segmentation"].plane_segmentations["roi_table"].to_dataframe()
roi_table.head()

there are 226 ROIs with soma_probability > 0.005

(256, 512)

(1736, 13799)

(226, 13799)

(226, 13799)

Text(0.5, 1.0, 'Spontaneous pre')

Text(0.5, 1.0, 'Spontaneous post')

(226, 226)

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

# Get the roi_table from the NWB file
roi_table = nwbfile.processing["processed"].data_interfaces["image_segmentation"].plane_segmentations["roi_table"].to_dataframe()
roi_table.head()

# Figure parameters for vertically stacked plots 
fig, ax = plt.subplots(figsize=(8, 4))

# Plot ROI contours 
for i, mask in enumerate(roi_table.image_mask):
    mask_array = np.array(mask)
    # Find contours at a level that works well for binary masks
    contours = measure.find_contours(mask_array, level=0.5)
    
    for contour in contours:
        color = 'k'
        linewidth = 1.0
        # Plot contour (note: contour coordinates are in (row, col) format)
        ax.plot(contour[:, 1], contour[:, 0], color=color, linewidth=linewidth)
        
# Label
ax.set_title('All ROI masks')

plt.tight_layout()
plt.show()

# Remove ROIs with traces that are NaNs (note - this takes a few seconds)
valid_trace_ids = [i for i in range(dff_traces.shape[1]) if np.isnan(dff_traces[0, i])==False]
# Limit ROI table to non-NaN traces
roi_table = roi_table.loc[valid_trace_ids]

# Find the likely somatic ROIs
soma_probability = 0.005 # Emperically determined threshold - just trust us
# Limit to valid somatic ROIs
valid_rois = roi_table[roi_table.soma_probability>soma_probability]
print('there are', len(valid_rois), 'ROIs with soma_probability >',soma_probability)

there are 226 ROIs with soma_probability > 0.005

# First load our average image
average_image = nwbfile.processing["processed"].data_interfaces["images"].images["average_projection"].data[:] 
average_image.shape

(256, 512)

# Plot ROI contours overlayed on average image 
fig, ax = plt.subplots(figsize=(8, 6))

# First plot average image
ax.imshow(average_image, cmap='gray', origin='upper', aspect='equal', vmax=0.25)
ax.set_title('ROI Contours Overlaid on Average Projection for Valid ROIs')
ax.axis('off')

# Overlay ROI contours
for i, mask in enumerate(valid_rois.image_mask):
    mask_array = np.array(mask)
    # Find contours at a level that works well for binary masks
    contours = measure.find_contours(mask_array, level=0.5)
    
    for contour in contours:
        color = 'yellow'
        linewidth = 0.5
        
        # Plot contour (note: contour coordinates are in (row, col) format)
        ax.plot(contour[:, 1], contour[:, 0], color=color, linewidth=linewidth)
        
# Set the axis limits to match the image dimensions
ax.set_xlim(0, average_image.shape[1])
ax.set_ylim(average_image.shape[0], 0)  # Flip y-axis to match image coordinates

plt.tight_layout()
plt.show()

# Check the shape of our dff traces array - the longer axis is 2P frames, the shorter one is ROIs
dff_pre.shape

(1736, 13799)

# Limit dff traces for spontaneous activity period before the task to valid ROIs
dff_pre = dff_pre[valid_rois.index.values, :]
print(dff_pre.shape)

(226, 13799)

# Limit dff traces for spontaneous activity period before the task to valid ROIs
dff_post = dff_post[valid_rois.index.values, :]
print(dff_pre.shape)

(226, 13799)

plt.pcolormesh(dff_pre, vmin=0, vmax=5, cmap='magma')
plt.colorbar(label='dFF')
plt.xlabel('2P frames')
plt.ylabel('ROI ID')
plt.title('Spontaneous pre')

Text(0.5, 1.0, 'Spontaneous pre')

plt.pcolormesh(dff_post, vmin=0, vmax=5, cmap='magma')
plt.colorbar(label='dFF')
plt.xlabel('2P frames')
plt.ylabel('ROI ID')
plt.title('Spontaneous post')

Text(0.5, 1.0, 'Spontaneous post')

correlations_pre = np.corrcoef(dff_pre)
correlations_post = np.corrcoef(dff_post)

correlations_pre.shape

(226, 226)

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

1.7649641007623317e-11

<matplotlib.legend.Legend at 0x7fe288d53970>

<matplotlib.legend.Legend at 0x7fe288dbae60>

284.7976594244874

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

correlations_pre = np.corrcoef(dff_pre)
correlations_post = np.corrcoef(dff_post)

correlations_pre.shape

(226, 226)

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

1.7649641007623317e-11

<matplotlib.legend.Legend at 0x7fe288d53970>

<matplotlib.legend.Legend at 0x7fe288dbae60>

284.7976594244874

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

correlations_pre = np.corrcoef(dff_pre)
correlations_post = np.corrcoef(dff_post)

correlations_pre.shape

(226, 226)

plt.pcolormesh(correlations_pre, cmap='RdBu', vmin=-0.1, vmax=0.1)
plt.colorbar(label='Correlation')
plt.xlabel('ROI ID')
plt.ylabel('ROI ID')
plt.title('Correlations pre')
plt.axis('square')

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

plt.pcolormesh(correlations_post, cmap='RdBu', vmin=-0.1, vmax=0.1)
plt.colorbar(label='Correlation')
plt.xlabel('ROI ID')
plt.ylabel('ROI ID')
plt.title('Correlations post')
plt.axis('square')

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

correlations_pre = np.triu(correlations_pre, k=1)
correlations_post = np.triu(correlations_post, k=1)

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

1.7649641007623317e-11

<matplotlib.legend.Legend at 0x7fe288d53970>

<matplotlib.legend.Legend at 0x7fe288dbae60>

284.7976594244874

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

correlations_pre = np.triu(correlations_pre, k=1)
correlations_post = np.triu(correlations_post, k=1)

plt.pcolormesh(correlations_pre, cmap='RdBu', vmin=-0.1, vmax=0.1)
plt.colorbar(label='Correlation')
plt.xlabel('ROI ID')
plt.ylabel('ROI ID')
plt.title('Correlations pre')
plt.axis('square')

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

plt.pcolormesh(correlations_post, cmap='RdBu', vmin=-0.1, vmax=0.1)
plt.colorbar(label='Correlation')
plt.xlabel('ROI ID')
plt.ylabel('ROI ID')
plt.title('Correlations post')
plt.axis('square')

(np.float64(0.0), np.float64(226.0), np.float64(0.0), np.float64(226.0))

# Run the stats! 
t, p_val = stats.ttest_ind(correlations_pre.flatten(), correlations_post.flatten())
print(p_val)

1.7649641007623317e-11

# Plot both distributions
plt.hist(correlations_pre.flatten(), bins=100, color='green', label='pre', histtype='step')
plt.hist(correlations_post.flatten(), bins=100, color='purple', label='post', histtype='step')
plt.title('Correlations before and after BCI learning, p='+str(np.round(p_val, 5)))
plt.xlabel('Correlation')
plt.ylabel('# cell pairs')
plt.xlim(-0.1, 0.1)
plt.legend()

<matplotlib.legend.Legend at 0x7fe288d53970>

def get_roi_centroids(roi_table):
    '''
    Iterate through the image masks in the ROI table, 
    get the indices corresponding to each ROI, 
    then take the average to get the x and y locations of the cell centroid. 
    
    Returns an array with dimensions (n_cells, xy_position)
    '''
    centroids = []
    for mask in roi_table['image_mask']:
        ys, xs = np.where(mask)
        x = np.mean(xs)
        y = np.mean(ys)
        centroids.append((x, y))
    return np.array(centroids)

<matplotlib.legend.Legend at 0x7fe288dbae60>

284.7976594244874

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

def get_roi_centroids(roi_table):
    '''
    Iterate through the image masks in the ROI table, 
    get the indices corresponding to each ROI, 
    then take the average to get the x and y locations of the cell centroid. 
    
    Returns an array with dimensions (n_cells, xy_position)
    '''
    centroids = []
    for mask in roi_table['image_mask']:
        ys, xs = np.where(mask)
        x = np.mean(xs)
        y = np.mean(ys)
        centroids.append((x, y))
    return np.array(centroids)

<matplotlib.legend.Legend at 0x7fe288dbae60>

284.7976594244874

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

def get_roi_centroids(roi_table):
    '''
    Iterate through the image masks in the ROI table, 
    get the indices corresponding to each ROI, 
    then take the average to get the x and y locations of the cell centroid. 
    
    Returns an array with dimensions (n_cells, xy_position)
    '''
    centroids = []
    for mask in roi_table['image_mask']:
        ys, xs = np.where(mask)
        x = np.mean(xs)
        y = np.mean(ys)
        centroids.append((x, y))
    return np.array(centroids)

<matplotlib.legend.Legend at 0x7fe288dbae60>

284.7976594244874

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

def get_roi_centroids(roi_table):
    '''
    Iterate through the image masks in the ROI table, 
    get the indices corresponding to each ROI, 
    then take the average to get the x and y locations of the cell centroid. 
    
    Returns an array with dimensions (n_cells, xy_position)
    '''
    centroids = []
    for mask in roi_table['image_mask']:
        ys, xs = np.where(mask)
        x = np.mean(xs)
        y = np.mean(ys)
        centroids.append((x, y))
    return np.array(centroids)

# Calculate centroids for each ROI 
centroids = get_roi_centroids(valid_rois)
centroidX = centroids[:, 0]
centroidY = centroids[:, 1]

<matplotlib.legend.Legend at 0x7fe288dbae60>

284.7976594244874

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

# Calculate centroids for each ROI 
centroids = get_roi_centroids(valid_rois)
centroidX = centroids[:, 0]
centroidY = centroids[:, 1]

# Now plot them
plt.plot(centroidX, centroidY, 'ko', alpha = 0.2, label = 'ROI centroids')
plt.xlabel('X position') 
plt.ylabel('Y position')
plt.title('ROI centroids for high probability cell somas')
plt.legend()
plt.legend(bbox_to_anchor=(1.0, 1.0))

<matplotlib.legend.Legend at 0x7fe288dbae60>

import math

284.7976594244874

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

import math

centroid_cell_1 = centroids[0, :]
centroid_cell_2 = centroids[1, :]

math.dist(centroid_cell_1, centroid_cell_2)

284.7976594244874

pairs_data = []
for roi_id_1 in range(len(valid_rois.index.values)): 
    for roi_id_2 in range(len(valid_rois.index.values)): 
        centroid_cell_1 = centroids[roi_id_1, :]
        centroid_cell_2 = centroids[roi_id_2, :]
        distance = math.dist(centroid_cell_1, centroid_cell_2)
        r_value_pre = correlations_pre[roi_id_1, roi_id_2]
        r_value_post = correlations_post[roi_id_1, roi_id_2]
        r_value_diff = r_value_post-r_value_pre
        pairs_data.append([roi_id_1, roi_id_2, r_value_pre, r_value_post, r_value_diff, distance])

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

pairs_data = []
for roi_id_1 in range(len(valid_rois.index.values)): 
    for roi_id_2 in range(len(valid_rois.index.values)): 
        centroid_cell_1 = centroids[roi_id_1, :]
        centroid_cell_2 = centroids[roi_id_2, :]
        distance = math.dist(centroid_cell_1, centroid_cell_2)
        r_value_pre = correlations_pre[roi_id_1, roi_id_2]
        r_value_post = correlations_post[roi_id_1, roi_id_2]
        r_value_diff = r_value_post-r_value_pre
        pairs_data.append([roi_id_1, roi_id_2, r_value_pre, r_value_post, r_value_diff, distance])

pairs = pd.DataFrame(pairs_data, columns=['roi_id_1', 'roi_id_2', 'r_value_pre', 'r_value_post', 'r_value_diff', 'distance'])
pairs.head()

plt.plot(pairs.distance.values, pairs.r_value_pre.values, '.', color='green', label='pre', alpha=0.5)
plt.plot(pairs.distance.values, pairs.r_value_post.values, '.', color='purple', label='post', alpha=0.5)
plt.legend()
plt.ylabel('Correlation')
plt.xlabel('Distance (pixels)')
plt.title('Activity correlation as a function of distance for cell pairs')

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

plt.plot(pairs.distance.values, pairs.r_value_pre.values, '.', color='green', label='pre', alpha=0.5)
plt.plot(pairs.distance.values, pairs.r_value_post.values, '.', color='purple', label='post', alpha=0.5)
plt.legend()
plt.ylabel('Correlation')
plt.xlabel('Distance (pixels)')
plt.title('Activity correlation as a function of distance for cell pairs')

Text(0.5, 1.0, 'Activity correlation as a function of distance for cell pairs')

Column	Description
id	data asset id
name	filename of data asset (raw)
subject_id	numerical id for animal subject
session_time	experiment date (%Y-%m-%d %H:%M:%S)
session_type	experiment identifier
genotype	subject genotype
virus	injected virus type
ophys_fov	field of view identifier
session_number	behavior training session number

	project_name	session_type	_id	name	subject_id	genotype	virus	date_of_birth	sex	modality	session_date	age	session_time	targeted_structure	ophys_fov	session_number
0	Brain Computer Interface	BCI single neuron stim	03f1b82e-4494-46bc-a959-50c4b14d13da	single-plane-ophys_731015_2025-01-28_17-40-57_...	731015	Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-IC...	pAAV-hSyn1-RiboL1-GCaMP8s-WPRE	2024-03-14	Female	Planar optical physiology	2025-01-28	320	17:40:57.996000	Primary Motor Cortex	FOV_04	22.0
1	Brain Computer Interface	BCI single neuron stim	b28d6321-1b1b-4110-87fb-0e19325c1dfe	single-plane-ophys_767715_2025-02-17_17-41-50_...	767715	Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-IC...	pAAV-hSyn1-RiboL1-GCaMP8s-WPRE	2024-09-15	Female	Planar optical physiology	2025-02-17	155	17:41:50.758000	Primary Motor Cortex	FOV_03	8.0
2	Brain Computer Interface	BCI single neuron stim	580448b2-4b93-4477-b6ac-6e2972792fe0	single-plane-ophys_772414_2025-01-27_15-57-05_...	772414	Camk2a-tTA/wt;tetO-GCaMP6s/wt	pAAV-CaMKIIa-ChRmine-oScarlet-Kv2.1-WPRE - 7413	2024-10-15	Female	Planar optical physiology	2025-01-27	104	14:23:26.693000	Primary Motor Cortex	FOV_01	5.0
3	Brain Computer Interface	BCI single neuron stim	c1be8449-7bef-4967-88f5-6a6ffc29e5a3	single-plane-ophys_772414_2025-02-06_16-47-44_...	772414	Camk2a-tTA/wt;tetO-GCaMP6s/wt	pAAV-CaMKIIa-ChRmine-oScarlet-Kv2.1-WPRE - 7413	2024-10-15	Female	Planar optical physiology	2025-02-06	114	16:47:44.152999	Primary Motor Cortex	FOV_02	13.0
4	Brain Computer Interface	BCI single neuron stim	15b395c0-1caf-4be4-a212-a13b9a4d1dfc	single-plane-ophys_772414_2025-02-10_11-15-26_...	772414	Camk2a-tTA/wt;tetO-GCaMP6s/wt	pAAV-CaMKIIa-ChRmine-oScarlet-Kv2.1-WPRE - 7413	2024-10-15	Female	Planar optical physiology	2025-02-10	118	11:15:26.609999	Primary Motor Cortex	FOV_03	14.0

	project_name	session_type	_id	name	subject_id	genotype	virus	date_of_birth	sex	modality	session_date	age	session_time	targeted_structure	ophys_fov	session_number
26	Brain Computer Interface	BCI single neuron stim	b8827d25-495f-46a8-9f33-ffb24da527a5	single-plane-ophys_731015_2025-01-10_18-06-31_...	731015	Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-IC...	pAAV-hSyn1-RiboL1-GCaMP8s-WPRE	2024-03-14	Female	Planar optical physiology	2025-01-10	302	16:46:51.981999	Primary Motor Cortex	FOV_04; FOV_04	18.0
27	Brain Computer Interface	BCI single neuron stim	85ff1fdd-3415-46a7-af9a-c77385a68463	single-plane-ophys_731015_2025-01-13_20-09-20_...	731015	Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-IC...	pAAV-hSyn1-RiboL1-GCaMP8s-WPRE	2024-03-14	Female	Planar optical physiology	2025-01-13	305	18:53:03.001999	Primary Motor Cortex	FOV_04	19.0
28	Brain Computer Interface	BCI single neuron stim	127a3e78-729c-4df7-bf34-1b9308939587	single-plane-ophys_731015_2025-01-24_20-00-44_...	731015	Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-IC...	pAAV-hSyn1-RiboL1-GCaMP8s-WPRE	2024-03-14	Female	Planar optical physiology	2025-01-24	316	18:41:22.550000	Primary Motor Cortex	FOV_04	20.0
0	Brain Computer Interface	BCI single neuron stim	03f1b82e-4494-46bc-a959-50c4b14d13da	single-plane-ophys_731015_2025-01-28_17-40-57_...	731015	Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-IC...	pAAV-hSyn1-RiboL1-GCaMP8s-WPRE	2024-03-14	Female	Planar optical physiology	2025-01-28	320	17:40:57.996000	Primary Motor Cortex	FOV_04	22.0
29	Brain Computer Interface	BCI single neuron stim	265c6a41-2783-4edd-b707-e4feaa77aad2	single-plane-ophys_731015_2025-01-28_17-40-57_...	731015	Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-IC...	pAAV-hSyn1-RiboL1-GCaMP8s-WPRE	2024-03-14	Female	Planar optical physiology	2025-01-28	320	17:40:57.996000	Primary Motor Cortex	FOV_04	22.0
6	Brain Computer Interface	BCI single neuron stim	0162d41c-613c-4215-b0aa-9690f85a9fda	single-plane-ophys_731015_2025-01-28_18-56-35_...	731015	Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-IC...	pAAV-hSyn1-RiboL1-GCaMP8s-WPRE	2024-03-14	Female	Planar optical physiology	2025-01-28	320	17:40:57.996000	Primary Motor Cortex	FOV_04	22.0
5	Brain Computer Interface	BCI single neuron stim	b9a4c361-66b0-4cd5-9392-75d116ef3385	single-plane-ophys_731015_2025-01-31_20-37-19_...	731015	Slc17a6-IRES-Cre/wt;Ai230(TIT2L-XCaMPG-WPRE-IC...	pAAV-hSyn1-RiboL1-GCaMP8s-WPRE	2024-03-14	Female	Planar optical physiology	2025-01-31	323	20:37:19.623000	Primary Motor Cortex	FOV_04	23.0

	start_time	stop_time	start_frame	stop_frame	tiff_file	stimulus_name	laser_x	laser_y	power	duration	stimulus_function	group_index	closest_roi
id
0	0.000000	0.617884	0	36	spont_slm_00001.tif	photostim	241.425926	212.703704	4	0.092	scanimage.mroi.stimulusfunctions.logspiral	78	421
1	0.635047	1.218604	37	71	spont_slm_00002.tif	photostim	168.200000	149.975000	4	0.092	scanimage.mroi.stimulusfunctions.logspiral	55	355
2	1.235767	1.836487	72	107	spont_slm_00003.tif	photostim	165.013514	133.905405	4	0.092	scanimage.mroi.stimulusfunctions.logspiral	8	153
3	1.853651	2.454371	108	143	spont_slm_00004.tif	photostim	239.032258	40.870968	4	0.092	scanimage.mroi.stimulusfunctions.logspiral	65	1169

	start_time	stop_time	go_cue	hit	lick_L	reward_time	threshold_crossing_times	zaber_step_times	tiff_file	start_frame	stop_frame	conditioned_neuron_x	conditioned_neuron_y	closest_roi
id
0	1692.228741	1704.020019	0.2359	True	[1.7861, 1.9033, 2.0241, 2.1415, 2.2464, 2.3756, 2.5235000000000003, 6.5394000000000005, 6.6744, 6.7902000000000005, 6.8994, 7.011100000000001, 7.134200000000001, 7.2574, 7.3948, 7.531000000000001, 7.6886, 7.843400000000001, 7.996300000000001, 8.1553, 8.368500000000001, 8.702, 8.8604, 9.0936]	1.7861	1.7089	[0.3535, 0.5054000000000001, 0.6074, 0.7184, 0.7994, 1.2374, 1.3464, 1.4334, 1.4964000000000002, 1.5824, 1.6234000000000002, 1.6604, 1.6924, 1.7244000000000002, 1.7564000000000002, nan, nan, nan, nan, nan, nan, nan]	neuron119_00001.tif	98595	99282	108.633333	51.666667	124
1	1704.037183	1708.722800	0.2359	True	[1.0667, 1.2012, 1.3251000000000002, 1.4520000000000002, 1.5643, 1.6967, 1.8427, 2.0133, 2.2004, 3.5486, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan]	1.0667	1.0582	[0.2869, 0.3589, 0.4188, 0.4969, 0.5359, 0.5779000000000001, 0.8398, 0.8718, 0.9038, 0.9358, 0.9678, 0.9998, 1.0318, 1.0638, nan, nan, nan, nan, nan, nan, nan, nan]	neuron119_00002.tif	99283	99556	108.633333	51.666667	124
2	1708.739964	1714.060628	0.2359	True	[0.7160000000000001, 0.8175, 0.9416, 1.0605, 1.1738, 1.3011, 1.4266, 1.5732000000000002, 1.7261000000000002, 2.182, 2.4106, 2.5404, 2.7478, 3.7818, 3.904, nan, nan, nan, nan, nan, nan, nan, nan, nan]	0.8175	0.7399	[0.3182, 0.3502, 0.3822, 0.4142, 0.4462, 0.4782, 0.5102, 0.5422, 0.5742, 0.6062000000000001, 0.6382, 0.6702, 0.7022, 0.7342000000000001, 0.7662, 0.7982, nan, nan, nan, nan, nan, nan]	neuron119_00003.tif	99557	99867	108.633333	51.666667	124
3	1714.077792	1718.488794	0.2359	True	[1.0345, 1.1526, 1.2687000000000002, 1.392, 1.5199, 1.6502, 1.8058, 2.173, 3.1982, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan]	1.0345	0.9577	[0.4984, 0.5503, 0.5874, 0.6204000000000001, 0.6533, 0.6864, 0.7193, 0.7554000000000001, 0.7993, 0.8313, 0.8633000000000001, 0.8953000000000001, 0.9273, 0.9593, 0.9913, 1.0233, nan, nan, nan, nan, nan, nan]	neuron119_00004.tif	99868	100125	108.633333	51.666667	124

Day 1 Workshop 2: Science Quest

How do neuron interactions change after learning?

SWDB 2025

Goal

Step 1: Which datasets have we learned about that would be useful to analyze to address this question?¶

Step 2: What analysis method can we use to address this question?¶

Let's use our new data access skills to load the BCI data and compare neuron activity correlations during the spontaneous periods before and after a mouse learns the BCI task!¶

Outline:

Part 1: Accessing the BCI data

Part 2: What are the conditions in this dataset?

Part 3: Quantifying activity correlations before and after BCI learning

Part 4: Do correlations depend on distance between neurons?

BCI dataset overview

Import packages and load data¶

Part 1: Accessing the BCI data

Metadata¶

Loading session data from an NWB file¶

root (NWBFile)

NWB files¶

Part 2: What are the conditions in this dataset?

Epoch Table¶

Overlay epochs with cell activity traces¶

Trials tables¶

Optogenetic photostimulation trials¶

BCI behavior trials¶

Could the learning that happens during the BCI task epoch change the activity of the other neurons in the imaging plane?¶

Part 3: Quantifying activity correlations before and after BCI learning

Woah thats a lot of ROIs. Are they all valid?¶

Image segmentation ROI masks¶

Limit spontaneous pre and post dff traces to valid ROIs¶

Quantifying correlations¶

Part 4: Do correlations depend on distance between neurons?

Calculate the centroid location for each ROI image mask and compute distances¶

Calculate distances between ROI centroids and relate to spontaneous activity correlations.¶

What other factors could influence how correlated neurons are during different epochs?¶

Questions and analyses to explore further:¶

	stimulus_name	start_frame	stop_frame	start_time	stop_time
id
0	photostim	0	84794	0.000000	1455.356193
1	spont	84795	90794	1455.373356	1558.336795
2	spont_01	90795	98594	1558.353958	1692.211577
3	BCI	98595	160009	1692.228741	2746.303855

Column	Description
start_time	stimulus start (s)
stop_time	stimulus end (s)
start_frame	stimulus start (frame)
stop_frame	stimulus end (frame)
tiff_file	data source file name
stimulus_name	stimulus name
laser_x	x coordinate of stimulated neuron (pixel)
laser_y	y coordinate of stimulated neuron (pixel)
power	stimulus intensity (mW)
duration	trial duration (s)
stimulus_function	stimulus template
group_index	identity of stimulated neuron(s)
closest_roi	index in dff that corresponds to the photostimulated neuron

Column	Description
start_time	trial start (s)
stop_time	trial end (s)
go_cue	time of go cue relative to start time (s)
hit	boolean of whether trial was hit
lick_l	lick times (s)
reward_time	reward delivery time (s)
threshold_crossing_times	time when reward port crossed position threshold (s)
zaber_steps_times	position of reward port
tiff_file	data source file
start_frame	trial start (frame)
stop_frame	trial end (frame)
conditioned_neuron_x	coordinate for conditioned neuron (pixels)
conditioned_neuron_y	coordinate for conditioned neuron (pixels)
closest_roi	index in dff that corresponds to the photostimulated neuron

Column	Description
is_soma	==1 if ROI classified as soma, ==0 if not
soma_probability	if >0.5 classified as soma
is_dendrite	==1 if ROI classified as dendrite, ==0 if not
dendrite_probability	if >0.5 classified as dendrite
image_mask	HxW sparse array defining image masks

	is_soma	soma_probability	is_dendrite	dendrite_probability	image_mask
id
0	0	0.000019	0	0.144048	[[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...
1	0	0.000021	0	0.000081	[[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...
2	0	0.219488	0	0.000000	[[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...
3	0	0.000000	0	0.000002	[[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...
4	1	0.999860	0	0.000000	[[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,...

Data type	float32
Shape	(249370, 1736)
Array size	1.61 GiB

Data type	float64
Shape	(3,)
Array size	24.00 bytes

Data type	float64
Shape	(2,)
Array size	16.00 bytes

Data type	float64
Shape	(3,)
Array size	24.00 bytes